Unresponsive on "Quick Search" stage with simple dataset about autoai HOT 5 OPEN

Can you try according to this as mentioned in the repo, I think the "features" part should not be the in the same code-line. Sample

from autoai.

@garrettjoecox, considering the size of the data, you might actually need GPU acceleration. To ensure that it is not stuck, can you try with just the first 1000 rows of the DataFrame?

from autoai.

Can you try according to this as mentioned in the repo, I think the "features" part should not be the in the same code-line. Sample

According to the docs this is valid, as I only want to train on those two columns

There might be scenarios where you want to explicitely exclude some columns, or only use a subset of columns in the training. Manually specify the features to be used. AutoAI will still perform a feature selection within the list of features provided to improve effective model accuracy.

model = bc.train(file="data.csv", target="Y_value", features=["col1", "col2", "col3"])

However since I have already stripped everything else from my dataset I don't need to specify them as they are the only columns remaining, so I tried removing the features argument and got the same result.

from autoai.

@garrettjoecox, considering the size of the data, you might actually need GPU acceleration. To ensure that it is not stuck, can you try with just the first 1000 rows of the DataFrame?

I tried with less data, (1000, 100, 50, 10 rows) and it seems to have gotten a bit further, but then the kernel dies every time around 30-40% of the way through this step:

from autoai.

Thank you for the update @garrettjoecox. My best guess is that it is crashing on one of the models. This problem is most likely data specific. Is it possible for you to email the first 1000 rows of data to [email protected]? Trying it on your data will allow us to diagnose better.

from autoai.